IE9 Preview Touts Cross Browser Compatibility

An anonymous reader writes "Microsoft's Internet Explorer 9 development team has announced the availability of the third IE9 platform preview release on the IE blog. Dean Hachamovitch writes, 'The third Platform Preview of Internet Explorer 9, available now, continues the deep work around hardware acceleration to enable the same standards-based markup to run faster. This is the latest installment of the rhythm we started in March, delivering platform preview releases approximately every eight weeks and listening to developers. You'll see more performance, same markup, and hardware-accelerated HTML5.' The announcement focuses on cross-browser compatibility, noting that when 'developers spend less time rewriting their sites to work across browsers they have more time to create amazing experiences on the Web.' Curiously, however, the video embedded in the page works only in some browsers. Dear Microsoft, IE9 supports many royalty-free, web-compatible formats out of the box (HTML, CSS, WOFF, PNG, and the like) so why not at least one more?"

Comments

Doesn't matter

[capnchicken]

I'll be writing shit web code for IE6 forever anyway.

Re:Doesn't matter

[capnchicken]

When one of your biggest clients happens to be GM or Blue Cross, it doesn't happen very quickly or ever. I think there are still quite a few Win 2000 machines over there.

Re:Doesn't matter

[betterunixthanunix]

I am not a web developer, so I am a bit confused about why websites are unable to provide even a basic level of support for IE6 -- perhaps a simple page without any fancy effects that just gives people whatever information they were looking for. Is it really that necessary to use Javascript for everything?

Re:Doesn't matter

[denis-The-menace]

Because IE 6 was designed to fail horribly when it can't understand the web page. Therefore web site developers have to make it work in IE or give it a way to fail nicely.

Re:Doesn't matter

[dagamer34]

Because that keeps enabling IE6. And providing even "basic support" takes a lot of work because of the number of hacks needed to make anything look decent. It's basically designing a different site just for IE6. No one likes doing that, which is why web developers want to explicitly not support IE6 to avoid that headache.

Re:Doesn't matter

[watermark]

IMO, with the existence of javascript libraries like jQuery or prototype, javascript isn't the issue. IE6 has so many bugs to keep up with (double margin, float bugs) that it forces you to create 1 1/2 websites (code that only runs on IE6 is the 1/2.) You tell it I want a a 6px border and place it 10px from the top and you can end up with a 12px border and 12px from the top. So you end up writing code to say, if its IE6, give it a 3px border and 8px from the top (knowing that it will double your 3px and add a few pixels to the latter.) So to answer your question, if we didn't write code specifically for IE6 (even without javascript) some sites would look so bad that they would be unusable in IE6. Stupid blue E.

Re:Doesn't matter

[Trevelyan]

If you still see websites as just online magazines then I guess you're somewhat stuck in the 1990's.

What you suggest is like asking: Why can't we make programmes also run on Windows 95, just without all the fancy effects of Aero?

Like it or not websites will become more interactive, even /. is using AJAX these days.
See also Google Docs, or http://www.jsdesk.com/

Re:Doesn't matter

[WrongSizeGlass]

We absolutely can provide 'basic' or better support for IE 6 but we make sure the entire site works the same in all browsers. The real questions is "how many man hours, and how much of our client's money, should we spend trying to handle IE 6 compatibility and functionality?"

Some of our sites are relatively simple and use formats & javascript libraries we've developed that support IE 6 without any issues at all. If a new or existing client wants this kind of site we don't "break" the IE 6 support - we simply let them know that any future upgrades or additions may not be able to do everything they want (either we can't add the 'fancy' stuff because it would require reworking of the infrastructure or it would break IE 6 compatibility). We always give them the option and if they are willing to spend the extra money on IE 6 compatibility we'll do it but we design the entire site to work the same in all browsers so they have a limited feature set to choose from.

We see dwindling IE 6 visitors but we're also aware that some of our clients still have a large percentage of IE 6 visitors. Because of that we don't offer certain features or enhancements to them. We're upfront about it and they are willing to live with the less fancy site as long as it meets the needs of their visitors.

Re:Doesn't matter

[imakemusic]

Because it could double your development time and therefore development cost.

I agree with you...most of the time. Most people should design their sites for standards compliance and leave as is for IE=6. You can use a "IF IE 6" tag to show a banner at the top of the site saying something to the effect of "This site doesn't look right. That's because your browser is a piece of crap. You can, however, continue to use this site. It just won't look very nice or work very well. Here's a link to list of reasons why IE6 should die in a ditch." This, I think, is the best way to deal with the problem.

This is all well and good if your site is relatively simple (as most are) but the problems come when you're making a site that relies heavily on something that has to be done differently for IE6, such as JavaScript. For example, I don't know if Google Maps works on IE6 but I wouldn't be surprised that, for something of that complexity, most of the client-side code would need to be written twice - once for the standards and once for IE.

When providing even a basic equivalent of your site for a tiny percentage of the sites users almost doubles development time it just might not be worth it.

(note: for most sites "double" is an exaggeration...but I think my point still stands)

Re:Doesn't matter

[Sleepy]

"I am not a web developer, so I am a bit confused about why websites are unable to provide even a basic level of support for IE6 -- perhaps a simple page without any fancy effects that just gives people whatever information they were looking for. Is it really that necessary to use Javascript for everything?"

Good question, but it's not that simple.

See, CSS and Javascript were DESIGNED to "fail gracefully". You could put some useful style on say a list or a heading, then use CSS to format it. If the browser support was not there, you would see the base elements.

Now this failback would understandably be UGLY (your prettified CSS list menu would look like a 1994 bulleted list), but it would WORK.

If Microsoft chose not to support the CSS standards, they could have done so. It's optional.

The ONLY way Microsoft could ruin CSS and Javascript was by agreeing to go along with the standard, and then change all the meaning. It's like if you spoke a slightly different language than your neighbor, and every 3rd word you spoke had different meaning to your neighbor (as in, every 3rd word was a normal term to you, but an unexpectedly offensive curse word to them).

If a browser did not support said standards, we could have all designed for CSS and IE6 would get a vanilla plain text page.

Microsoft knew that novice web developers would code and test in the "popular" browser first, then test other browsers afterwards. If that was how you developed, you were an unwitting tool in Microsoft's effort to destroy the open web.

It worked, for a time.

Then web developers revolted, by figuring out how to document Microsoft bugs. In the end, we developed this pseudo-language that ran on top of CSS and Javascript, so we could "hide" markup and styles from either IE or from the standards browsers.

All this effort wasted uncountable hours of web developers.

Was this deliberate sabotage by Microsoft? Let's just say that in the US anti-trust trial against Microsoft, emails from Bill Gates were revealed. Bill's emails essentially stated he didn't want to see MS developers "wasting time fixing bugs in HTML that only affect competitive browsers". (Meaning, if your HTML/CSS generation in some desktop app generates horrible invalid code... DON'T fix it... just let the IE guys know so they can write undocumented code to show your page "properly").

Literally, there's a story here how grass-roots web developers fought back to save the "information highway" from being effectively privatized as one company's property.

This is why so many people HATE IE and IE6.. even if they're not the type of people who normally hate Microsoft.

Re:Doesn't matter

[Rockoon]

The real questions is "how many man hours, and how much of our client's money, should we spend trying to handle IE 6 compatibility and functionality?"

Thats for the client to decide.

Its as if you think that if IE6 was eliminated from the landscape, that you would get to do the job in less time for the same money. Thats not how markets work. If your job gets easier, so does your competitions job.

Re:Doesn't matter

[BenEnglishAtHome]

If you still see websites as just online magazines then I guess you're somewhat stuck in the 1990's.

I am, I suppose, well behind the times. Some years ago, websites were able to publish information and update the databases they sat in front of (to do things like take orders for merchandise or carry on a conversation), all over secure connections. Thus, in the beginning the web established usefulness as a publishing medium. Then it was a tool to get things done.

Since then, pretty much everything I've seen in various "WebX.0" applications has done the same stuff in different ways.

So...beyond what we had years ago, what sorts of "interactivity" actually serve a useful purpose?

I'm not trying to be an obtuse old man. Seriously, I'd like to know. Maybe there's some really useful thing that I should be doing via the web that I'm not doing now. But I can't really think of what that might be and I'd like some help to figure it out. What new interactivity justifies the visual pollution and system crud that are (apparently) required to make use of it?

Note - "Mobile" apps don't count. I can understand why I might need some specialized program when I'm trying to get something done while moving around. I haven't bought into the "I live through my smartphone" lifestyle, though. I'm asking about things that are useful to me when I sit down at my desktop computer.

Re:Doesn't matter

[kimvette]

I'd like to invite you to check out the lynx and links web browsers.

The problem with MSIE6 is that it adds nonstandard extensions to HTML and CSS, does not (natively) support the full PNG spec, it is pathetically insecure, it adds padding to certain HTML elements in a lot of situations where everybody else assumes padding=0 so by making a web page in standard HTML/XHTML that looks gorgeous in every single other web browser will be horribly broken in MSIE6.

MSIE7 and MSIE8 have progressively gotten a TON better but they still don't handle broken pages gracefully (see the acid3 test) and will still degrade to MSIE6 compatibility mode, encouraging corporate web developers to be lazy and keep things as-is.

Now, as far as "fancy effects" go - those "fancy effects" led to the possibility of google docs, web-based photoshop elements replacements, online banking that doesn't take weeks to navigate (do you have an AmEx account? Log into your account on AmEx and you will see online commerce done right), and even legal free and low-cost on-demand video programming, and online classes. It also allowed for the building of "community" web sites that made the whole world a lot smaller, connecting people from nearly every nation.

It also allows us to research products better before we buy, so when companies post their products online they can post a LOT more detail than they ever did in printed brochures, and you can educate yourself so that salespeople who nothing about the product beyond what their commission is won't steer you wrong.

You can stick to the web as it was circa 1997. I'll take today's "web 2.0" (wait, did I just say web 2.0?), thankyouverymuch. And, I'll be very happy using non-Microsoft browsers.

Re:Doesn't matter

[RobertM1968]

Actually, there was no designing involved in the failing parts. The code just ran into unspecified territory. You could say the outcome was determined by natural selection. ;)

Or... you could say it was designed into it in many areas where Microsoft intentionally ignored standards and inserted their own method of doing things. I would agree with you, if it weren't for all the memos, emails and such that went back and forth at Microsoft about trying to kill Netscape dominance by doing such things (see the DOJ case against MS for more info).

While the PP may seem like a conspiracy theorist, the fact is, ensuring web code would fail on IE unless written to Microsoft's specs/"standards" was indeed an intentional move on their part, combined with their push into the business world for their various technologies that only worked on IE. The intent, as I noted above, was to ensure that they would gain more marketshare over Netscape and take the Internet playing field away from them. They succeeded, and hence, we have IE6 (and IE7 and IE8) to code workarounds for.

I for one still remember the day of going to websites that were IE6 only...

Re:Doesn't matter

[Hal_Porter]

Yeah, I don't get it either. A long time ago I worked on a web app that supported IE3, IE4 and Netscape. All the layout was done with tables. Validation was done both on the server and in JavaScript. We sniffed the browser and turned off JS if the browser was not whitelisted. The site would still work - it just needed to round trip to the server if a form contained an invalid date for example. On a browser with decent JS support you could skip that step and reduce the server load a bit.

IE3 was very quirky, IE4 wasn't too bad and Netscape got most stuff right even though it was a bit old. But the corporate desktops we were aiming at demanded IE3 compatibility so we made sure that we didn't need much sophistication on the client.

Why it's "OMG totally impossible" to support IE6 is frankly beyond me.

Now accelerated to pwn your machine 100% faster

[Rogerborg]

With smoother and more convenient Ring 0 integration than ever!

Re:Cross Browser Compatibility?

[The MAZZTer]

Browsers can be made to be standards compliant. Web pages can be made to be cross-browser compatible (since not all browsers are standards compliant).

Re:Cross Browser Compatibility?

[capnchicken]

Nope, they don't want to be tied down to such non-marketable terms such as "standard" or such constricting terms like "compliant".

"Cross Browser" sounds WAY sexier and "compatible" sounds much less like they HAVE to do something.

Re:Cross Browser Compatibility?

[starfishsystems]

Nah. Microsoft seems pathologically unable to acknowledge the possibility of an objective standard.

A few browsers which more or less by accident behave similarly, now that's a vision that Microsoft can get behind! That situation can be manipulated. Objective standards, on the other hand, are the enemy of relativism.

Whoah that's about time, i'd almost believe it..

[SchizoDuckie]

Too bad that the IE example doesnt properly in Chrome because it *requires* hardware accelleration (if that's either to their crappy javascript or the amaaaazing speedup they finally got working i'll leave in the middle) The Good thing though: It really works! I put it through it's paces with Peter Nederlof's (A.k.a. Clay) 3d javascript engine to see what part is crappy and what's working, and for now it looks AWESOME! The only thing that doesn't seem to work is click tracking on the canvas. speed wise it's quite similar to Chrome! Test urls: http://www.xs4all.nl/~peterned/3d/ http://www.xs4all.nl/~peterned/demooo/duck.html http://www.xs4all.nl/~peterned/demooo/cubes.html

Re:Microsoft supports H.264

[PerfectionLost]

H.264 is still a proprietary codex.

Re:Cross Browser Compatibility?

[Richard_at_work]

Even when following standards to the letter, not all browsers produce the same result - there are plenty of examples where both Firefox and Safari 'get it right' according to the standard, and yet produce different results. Accommodating these sorts of things is also covered under the term cross-browser compatibility.

Re:Cross Browser Compatibility?

[IntlHarvester]

No, "cross browser" accurately describes what people want. Nearly always, when some internet nerd starts whining about "standards-compliance", they no clue what the standards actually are, and what they really mean is "Make it work like Firefox".

Realistically, there are hundreds of "standards" which no browser supports, and there are numerous de facto unofficial standards which people expect to work. (Prime example of the latter is transparent PNGs.) "Cross-browser" accurately describes this set of common pratices.

Re:Cross Browser Compatibility?

[IntlHarvester]

Actually, Microsoft is generating a tons of CSS test cases, which more of an objective measure than the checkbox marketing which usually is used by browser vendors.

Without Firefox...

[Hurricane78]

Did anyone else think that we really have to thank the Mozilla team for this? Without Firefox, none of this would have happened. Wed’d still use IE6.

Firefox tends do go a bit downwards in quality, lately. But I don’t care. Thank you, Mozilla team! Every single one of you. Everyone who installed and promoted it. And the team who made the great logo and CI, that’s so fashionable that non-geek women put in on their t-shirts.
*grabs web-Oscar, steps down from the podium and runs away with it!*

Re:Cross Browser Compatibility?

[IntlHarvester]

(Apologies for replying to myself.)

The biggest problem when discussing web standards is that the vendors themselves propose the standards. So Apple is the most compliant with Apple's proposed standards, while Microsoft is the most compliant with Microsoft's proposed standards, etc. From the W3C's POV they are all the same, while the marketplace sorts these things out into common "cross-browser" features versus things which are considered "proprietary".

In other words, nobody cares that CSS3 rounded borders aren't an official "standards compliant" feature, it is a "cross browser" feature and they want rounded fucking corners on their website.

HTML5 Canvas Support

[butlerm]

The best part about this preview is the addition of HTML5 Canvas support, the lack of which would be a serious impediment to cross platform deployment of a large number of useful applications.

Re:Microsoft supports H.264

[Hurricane78]

And still, nobody cares. ;)
The specs are available. There are open source codecs (=encoder/decoder) for it.
And nobody cares what MPEG LA wants.
They gave the information out, and did not demand something in return. Now it’s too late. <stew-beef>MPEG LA, go fuck yourself!</stew-beef> ;)

Re:Relevance

[innocence18]

I've always wanted to say this....you must be new here

IE9 and WebM

[VGPowerlord]

I saw another article over at The Register about the new IE9 preview.

There was one section I found particularly interesting in there:

And speaking of standards, director of IE product management Roger Capriotti told The Reg that although patent-backed H.264 is Microsoft's video codec of choice for IE9, it will also support Google's recent open source gift to the world, WebM/V8, if a user has that codec on their machine. However, Redmond has yet to make a decision, Capriotti said, about how to handle the open source Ogg Theora codec.

So... IE9 will support WebM if it's installed, but not Theora.

While this is not supported out of the box, this could actually be a tipping point for WebM.

Without IE9's WebM support, things looked like this:

H.264 support: IE, Safari, Chrome
WebM support: FireFox, Chrome, Opera

In that case, H.264 looked like the winner. But if you add IE9 to the WebM column, you suddenly have support for WebM from everyone but Apple.

Now the trick will be to convince MS to support this out of the box...

Reminds me of a situation I had with a new phone..

[SharpFang]

back then I was scrapping for money, camera phones were relatively new, I needed a digital camera and couldn't afford one, and I could get a phone with decent camera, with a contract, for very reasonable money. And I needed a new phone anyway.

So I picked one. It could make photos okay, but to get them I could only send them through MMS to my email, for exorbitant fees. To download them I needed a special RS232-based cable... and the dealer didn't have them. No import, not available, if ordered from the net, including shipping, it would cost more than the phone, and about as much as a digital camera. But hurray, there are cheap chinese USB cables that supposedly work!

And they do, for everything EXCEPT downloading the photos. A 3rd party app can download thumbnails of the photos. The official app doesn't recognize the cable. The fora are filled with people asking how to get the photos, the universal answer is "get the official cable".

Quite pissed off, I first hacked together a RS232 cable using the plug from the chinese one and a handful of electronics. I found out the only difference from the "unofficial RS232" was that official had DTR and RTS shorted, the knock-off - unconnected. Still not satisfied I began reverse-engineering the AT command set the phone used to talk with the computer. I found commands to request list of photos, download and delete them, then how to extract the photo from the junk the phone sends as reply to request... I wrote a Perl app that worked with any serial, even the emulated RS232 over USB. It was clunky, it worked from command line only, but it worked with any cable.

I posted it on the official fora. To my surprise, instead of ban&delete, I received a surprised question from the developers: Why? Why would anyone want to use it? We have the official app which is infinitely better!

I explained how there are no official cables in my country. How I bought a phone for the camera, and I can't use the camera. That I understand they want to profit from their cables, but sorry, I feel cheated, I want to use the camera. Oh, and I listed an extract from first page of the support forum, about 20 posts of cable problems, to which my program was a solution.

That was the last I used my app. A new version of the official app was released less than a week later, and it ignored the DTR/RTS, working correctly with all cables.