This is the personal blog of Chris Wilson, Platform Architect of the Internet Explorer Platform team at Microsoft (and ex-Group Program Manager).
If you would like to receive an email when updates are made to this post, please register here
Subscribe to this post's comments using RSS
Congatulations!!!
Seriously. Dude. I was NOT expecting this. Guess we finally discover the/a reason for the delay on any news, you guys were waiting to drop the bombshell!
Thank you so much! You DO care!
First of all, congratulations. I didn't think I'd ever see the day that MSIE passed the Acid2 test. Smiley faces all round!
Second, the bad bit. Why don't you tell us what you are up to? Why is everything shrouded in secrecy? You would get a lot less flak if you gave developers a tiny clue as to where you are headed. It seems that you are doing some good work on the IE project team. Share a little of that with us and you won't get beaten up so badly in the blogs. ;-)
Congrats again. Really.
We love you, man, in a completely non-stalker sort of way.
Yeah, I agree with Dean Edwards: what's up? I'm an ASP.NET developer, so I do both server side .NET development and client side HTML and CSS. There is so much more information, excitement, positive development in/around .NET then there is for IE. It really seems like the whole .NET and so called Developer Division is like the 'new' Microsoft while IE is still very much the 'old' Microsoft.
Having said that (in my previous comment) I think IE is not part of the developer division, is that right? The latest post on the IE blog also shows that your are not using Team System (at least not for source control, I can see that from the //depot path, which looks like Perforce). If you use Team System, you might be able to open up a project on Codeplex to gather developer feedback in the form of work items etc.
Looking forward to your reactions!
Chris, that's a great news!
Are you guys planning to work with front-end developers like you did with IE7? I feel that would help a lot!
Hey Chris, I posted thanks on the IE blog already but it's worth passing on again. Hopefully this will really help push the web forwards. This is an excellent early Christmas present.
The problem though is the rate of update of browsers. IE7 is only now pushing over 50% of IEs for us, and IE6 isn't going to go below our support spectrum for a long time (for us this would be somewhere below around 1% of visitors). Unlike Firefox (which has an excellent updating system to the point that we can usually drop older versions within 6 months) IE is tied to the OS and can only be updated via OS updates. So I suppose my two questions are:
1) Is IE7 going to be a part of XP SP3?
2) Is IE8 going to be available for XP?
A yes to either of those will be good; a yes to both of them would be fantastic.
Great Christmas present. Glad you spent time on this rather than wasting time responding to continual pointless comment posts on the IEBlog.
@Dean + Mike
Just one man's opinion here:
http://www.d-boss.com/comment/jm-on-ie.htm#101207
Firstly, this is fantastic news. Great credit should go to yourself and the IE team for this. Award pats on the back all round. But don't sit on your laurels either, let's have more good stuff.
Secondly, can you clarify "IE8 standards mode"? Is this to be triggered in the normal manner - e.g. valid code with a valid doctype and so on - or do you have to do something IE8-specific to switch it on (in which case it's hardly complying to a *standard*).
But well done for recognising the importance (and high profile of this). Please, please, keep regular updates coming: it may be difficult to be transparent (particularly when you get so much negativity), but this is the best way to show us MS are listening.
Congratulations. However, I hope you are aware of the recent server problems of the Wasp-Server. Would be a shame if you were testing with a broken test.
Well, sorry for bad-thinking. I didn't expect this. I can't wait to test a beta :)
@JackP: From Chris's previous posts, comments and interviews it should be very clear, that IE8 will get a third rendering mode. This is neccesary for backwards compatibility. It is unclear how, but there's a distinct probability that web devs will have to opt-in to the new mode (that's actually great, since currently available websites won't break as much as they did when IE7 was released:)).
Thanks so much for keeping the community updated. This stuff matters; trust has to be regained. Looking forward to big things in IE8.
This is great news. Of even greater news I believe is that hasLayout is history in IE8. I wonder what other surprises are coming?
It's awesome news and you of all people should get to tell us ;)
Like many others I don't quite get why it had to be kept secret, but it's a clear signal that IE is getting so much closer to the other browsers - we can start getting excited ;)
Hi Chris,
Nice news! I'll merely repeat here a comment I left on the actual announcement. Just know that I've read the exchanges you had with some W3C WG on the matter.
About the IE8 Standards mode opt-in switch: I wonder about a few things...
- if IE 8 specified what file types it actually supports (like mentioning text/html, and application/xhtml+xml when it does), and gets content encoded as application/xhtml+xml, it should be able to directly use IE8 standard mode;
- if received content doesn't actually validate (meaning, content doesn't match DOCTYPE: MARQUEE inside HTML 4.01 Strict, etc.), automatically parse using 'older' IE7 engine; Opera does something like this on XHTML documents ('parse as HTML' in case of errors), and show a small icon in the corner (like IE does now in case of problems with Jscript).
What does it mean? Simply put, if someone has taken the trouble to create a page that validates on Strict DOCTYPEs (or uses a document type that IE7 and lower don't handle), there are good chances it will render well with IE8 'Standard' right out of the box.
By the way, whose idea was it to say 'IE can accept any file type' and make HTTP_ACCEPT useless with '*/*'? Couldn't you (the team) have at least used '*/*,q=0.1' and added 'text/html'? I'm sorry to say, but that was dumb.
I think that the best opt-in method you could use here is actually enforcing validation. The advantage of this method is that IE 8 will actually work like 'alternative' browsers: hacks like Holly Hack, the star CSS hack, browser object detection etc. would be handled like Opera or Firefox currently handle them.
Careful web designers will ensure that their websites don't trigger errors: don't forget to add a notification when IE 8 switches to 'legacy IE6/7 standard mode' (for Quirks, it's not needed - DOCTYPE switching works well enough).
Um... if IE8 requires a specific "I really want it to be rendered according to standards, honest guv, yes I do" mode that needs to be specifically triggered, then IE8 DOESN'T pass ACID2. It might render a hacked-up version of the ACID2 page successfully, but that's not the same thing.
And that's rather unfortunate, since you've gone to all the trouble to tell people you DO pass ACID2.
That's why I think it is critical for IE8 to trigger new "proper standards mode" in the same way that the previous standards mode was triggered - using a proper DOCTYPE and so on. Or at least by some other method WITHIN current standards...
Otherwise IE is continuing to propagate the non-standard route after all...
And let's face it, if someone has gone to the trouble of building a DOCTYPE and valid code, then they probably DO want their page to render according to standards. Don't make us developers jump through even more hoops. Please.
@JackP: Well, meta-tags aren't so clearly defined, that them opting-in to IE8's latest standards mode isn't really anything standards-harming.
However, if I heard right, this may not be the final solution, we'll just have to wait.
And come on, Introducing a new rendering mode is actually not worse. It has so many benefits!
* Websites that are working now will work in IE8 as well.
* New possibilities that are only in IE8 will bring competition to webauthors leading the web into a bright future.
Chris, I think you really do a great work. I've watched the recent interview on Channel9. It just made all my sorrow go away. Hope we can try this new beast very soon ;)
Meta tags I could live with! Particularly if it was some sort of meta tag that wasn't IE8 specific and therefore could be picked up on and used by other browsers.
What I don't want to have to do are insert conditional comments and the like everywhere...
Oh and yeah, Chris, well done and all that. Didn't mean to be adding to the negativity!
I don't care how it looks as long as is won't disturb standard based code.
Frankly, other browsers shouldn't include this meta-tag. For them, the doctype switch is more than enough. Don't forget: We aim to remove separate engines, not adding many more.
The meta-tag-solution is neccesary for IE, but to introduce it within other browsers would only cause more confusion.
Hands up - how many of you have put IE-specific hacks into your production code to make your pages work properly in IE - thereby EXPECTING IE to be broken?
Here's my point along the lines of what JackP is saying.
If you are going to "imply" that IE8 passes the Acid2 test when, in fact, IE8 DOES NOT PASS THE ACID2 TEST, and only "renders" the smiley face correctly, then this is deception, sleight of hand, smoke and mirrors.
You are advertising and hyping your product in a way that deceives your customers. And that is the problem I have.
This passing out of the box thing is not that important to. What matters is that I can actually choose which mode to use.
Chris, just one question. Are you going to freeze the two old modes, so even the version vector for Conditional Comments will stay the same in them?
You probably know the problem here. If I wrote for IE 7 and now there's IE8, the CC won't be used in IE8, although we're still in the old mode. Can you shed some light on this, please?
Oh, a bit too late, but .. happy new year :)
Chris Wilson said:
"Hands up - how many of you have put IE-specific hacks into your production code to make your pages work properly in IE - thereby EXPECTING IE to be broken?"
Myslelf for one. We expect IE 7, IE 6, IE 5 or earlier to be broken, not IE8.
Daniel wrote:
"The meta-tag-solution is neccesary for IE, but to introduce it within other browsers would only cause more confusion."
No it's not a solution at all, the solution is Standards out of the box.
http://lists.w3.org/Archives/Public/www-style/2007Dec/0148.html
And one solution to achieve that is here.
http://lists.w3.org/Archives/Public/www-style/2007Dec/0191.html
Re: "Hands up" - yes, both of my hands are WAAAY up!
I KNOW that:
IE5 is broken (but I don't support it anymore)
IE5.5 is broken (ditto)
IE6 is broken (serving up a bunch of hacks)
IE7 is broken (severing up less hacks)
IE8 (waiting to hear what the deal is)
As for the hacks that I'm serving up, 95% of them are in JavaScript, based on conditional statements, based on variables indicating which browser I'm working with.
I have an "IE" variable, and an "IE7OR_BETTER" variable.
when I want to float a div over a select, in goes the is(!IE7OR_BETTER){...} test to slip in an iframe under the div.
If IE8 ships, with everything just peachy, then I'll adjust so that IE8 gets pure juicy DOM scripting and I'll be happy.
If IE8 ships, with half-baked standards support, I'll likely add another variable into the mix, to only apply the hacks needed.
REGARDLESS, I want a FULLY standardized browser! Because the sooner that target is reached, the sooner I can write pure, clean code, and only sprinkle in the hacks for legacy IE browsers, using Conditional Comments and the like.
@Alan Gresley:
Yeah, standard out of the box'd be great. However that' utopical yet.
However, I truley believe with IE8 competition among webdesigners will grow in a direction where standards are inevitable. So I think IE8's mode is the best solution for today while "standards out of the box" (I read those mail archives btw), are the one and only long term goal.
@Webhosting, Vista, et al -
We're still thinking about freezing conditional comment version vector (and UA string). My gut instinct tells me we should leave the strategy alone there (i.e. even in IE7 mode, IE8 would have an IE8 UA string and an IE8 version for conditional comments), because code will change (for security reasons, etc).
@Chris
IE8 has Conditional Comments? Why? IE8 is an layout engine without hasLayout. What happens when IE8 in standard mode encounters this.
<!--[if gte IE7]>
<style type="text/css">
#container {height:1%}
</style>
<![endif]-->
How many sites will now break in IE8? How many pages does a developer have to change because they used this type of hack in the HTML?
But will it do so when released? The META tag enabling IE8 rendering isn't part of the Acid2 test. So how can it pass?
Not without some silly extra tag it won't.
Like many others I don't quite get why it had to be kept secret, but it's a clear signal that IE is getting so much closer to the other browsers - we can start getting excited ;)thank you!
The problem though is the rate of update of browsers. IE7 is only now pushing over 50% of IEs for us, and IE6 isn't going to go below our support spectrum for a long time (for us this would be somewhere below around 1% of visitors).
Art, I think you are totally right here
After I read this, I downloaded the IE8 beta and tried for myself. Chris Wilson is a liar, IE8 still failsails what in the name of.... My comments are overflowing the comment box and I can't read what I'm typing!?!?
Not only is he dishonest, he also writes very bad comment forms!
Not that you're a bad guy Chris, you're a great guy, and I love your work lately with the WHATWG and W3c and company... But very simply, IE8 doesn't pass Acid2, that's just rubbish.
Keep up the good work!
Congratulation and many thanks not only for this one but for your entire blog. Finally, there´s light at the end of the tunnel. I almost despaired with IE8!
Thank You Very Mach... CC IE8 olarak kullanılmayacaktır. Bu, lütfen bir ışık tuttu miyim?
