Internet Explorer 8 Beta: The Features: Internet Explorer 8 simplifies the use of standards-compliant XML-based webpages that support namespace features like scalable vector graphics, XML user interface language, mathematical markup language, and others.
To see the above text, click on “Faster, Easier”. Intuitive, I know.
Now lets try another version of that page which is served as text/html to IE, again with embedded SVG, and marked as X-UA-Compatible: IE=7. Whereas previously, it wasn’t all that compatible, IE8b2 at least attempts to be more compatible (complete with the invisible h1), until it gets to the SVGized W3C logo. As SVG images go, you can’t get much simpler than this: two consecutive path statements. But apparently, that’s beyond the limits of what IE8b2 supports, and it basically gives up parsing markup on the rest of the page. Converting the self-closing path elements to use explicit open and close tags works around the issue.
Finally, lets try something really simple. No SVG. No X-UA-Compatible. Served as text/html. The result: no CSS support for elements introduced in HTML5. While one can coax such support out of IE7 — and IE8 will even emulate that behavior when told to do so — I’m not aware of any equivalent mechanism for IE8.
Note: none of these pages were adjusted in any way to make IE8 look bad. They existed well before the beta was made available, and today was the first day I had access to IE8b2.
Just so I’m sure I understand:
<ol>
<li>XHTML not supported in IE8 (already knew this)</li>
<li>IE8 in Standards Mode no longer allows the createElement hack?</li>
<li>Inline XML has to have explicit closing tags?</li>
</ol>
I don’t believe that I have enough data to draw the third conclusion. In particular, the SVG icon that I’ve included on this post also consists exclusively of two self-closing path elements, and doesn’t seem to cause IE8 any problems.
Scott: can you describe what problem you are seeing? Firefox 3.0.1 is my primary browser, and I see no problem.
hdh: I’ve now cropped the screenshots. As to Microsoft’s level of understanding of the technologies they purport to enable... that’s something you will have to take up with them.
Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.0.1) Gecko/2008070208 Firefox/3.0.1
I’m not seeing the problem.
The symptoms are consistent with that page not being served as application/xhtml+xml, as the default behavior of at least two browsers (Firefox and Opera, IIRC) would not see the closing svg tag as closing all the intervening “self-closing” tags.
But you haven’t mentioned seeing that on my main blog — where I do the same thing, and try to do content negotiation. And that wouldn’t explain you seeing something odd on the third test.
But you haven’t mentioned seeing that on my main blog — where I do the same thing, and try to do content negotiation. And that wouldn’t explain you seeing something odd on the third test.
I do see it on the main blog as well as the Rails version. I just tried switching to a different proxy here at work, and that seemed to clear up the problem. I’m thinking that the proxy I’ve been using is mangling the content type. Further testing and/or complaints to proxy vendor will ensue. :)
Sam, I’m probably dense from tech editing, but I can’t see the text you reference, or even the link to click to get to the text. Is it still there? If so, I’ll have a cup of coffee and look again.
Update: According to the IEBlog, IE8 intends to support both; the difference being that IE=EmulateIE7 considers the HTML5 DocType to be a “quirks DOCTYPE” and will inexplicably no longer recognize foreign markup in such pages the way that IE7 used to.
Look at Tools -> Page Info. Type should be application/xhtml+xml.
I just got around to confirming the issue that I mentioned above. I am definitely seeing ‘text/html’ though the proxy where I saw the strange formatting. And I’m definitely calling that proxy vendor on Tuesday to complain, 'cause I had already asked the proxy to not mess with my Content-Type headers.
How odd. When I access the bug page with Firefox, I can see the page. When I access the bug page with IE8b2, I get an error that I don’t have permission to access the page.
I’m still trying to find my way through the convoluted patterns of access that IE8 implements. What am I doing wrong?
I think I answered my own question. I had a LiveConnect account to turn in bugs on Expression Media, and when I access the bug page, it remembers me and asks if I want to sign out.
I try, and I also deleted cookies, et al, but Microsoft does not actually delete the LiveConnect cookies, and that LiveConnect cookie is keeping me from seeing the bug in LiveConnect.
So no matter what I do, I’ll never be able to see this page, or get rid of the cookies Microsoft itself set.
My first experiences with IE8 beta 2 have been mixed. On the one hand, I like the fact that compatibility mode no longer requires restarting the browser. However, I’ve found it virtually impossible to tell when I have compatibility mode turned on...
