It’s just data

Improved Namespace Support

Microsoft: Internet Explorer 8 offers Web developers the opportunity to write standards-compliant HTML-based Web pages that support features (such as SVG, XUL, and MathML) in namespaces, provided that the client has installed appropriate handlers for those namespaces via binary behaviors... See Improved Namespace Support white paper for more information.

If I read this correctly, it should be possible for IE-8 to display the SVG that exists on my web page, assuming that the necessary handler is installed and deployed.  Microsoft may not be the one providing such a plugin, they are merely asserting that it can be done.

Besides the obvious (namely the fact that there is no standards compliant way to embed SVG in a non-XHTML page), this is very good news.  I hope other browsers (and, for that matter, the WHATWG) follow Microsoft’s lead.


Let’s hope that the now-EOL’ed Adobe SVG Viewer still continues to work in IE8... I’ve read that the plugin is also not supported on Vista.

Posted by Jeff Schiller at

Wow - the whitepaper is now up.  It’s better than I thought.  They use the inline SVG with HTML as an example:


<html>
<head>
  <title>SVG embedded inline in XHTML</title>
</head>
<body>
  <svg width="600" height="300" xmlns="http://www.w3.org/2000/svg">
  <linearGradient id="gradient">
  <stop style="stop-color:yellow" offset="0%" />
  <stop style="stop-color:green" offset="100%" />
  </linearGradient>
  <rect x="0" y="0" width="100" height="100" style="fill:url(#gradient)" />
  <circle cx="50" cy="50" r="30" style="fill:url(#gradient)" />
  <circle cx="150" cy="100" r="50" />
  </svg>
</body>
</html>

Btw, your formatting rules lie - white space is not preserved ;)

Posted by Jeff Schiller at

Significant restrictions per the whitepaper:

For my usage, that would affect any use of xlink.  I would be curious to find out what “not allowed” means, and would prefer to see the first three restrictions relaxed.  If “not allowed” means ignored, and future versions of IE must live with these restrictions, then that would be unfortunate.  Alternatively, if XHTML with the correct mime type is supported, and these are merely transitional aids, then that might be reasonable.

P.S.  Jeff: try putting code inside of triple curly braces.

Posted by Sam Ruby at

Came to the same conclusion here about xlink - but I’m not entirely convinced it wouldn’t work (since you’d declare the namespace at the top (the <svg> element).  If it doesn’t work, we should lobby to get that fixed in Beta 2 as xlink is a crucial namespace for SVG content.

Posted by Jeff Schiller at

You would have to declare the namespace at the top of the document, and then you couldn’t use it as an attribute.  The first is inconvenient, the latter is a showstopper.

Posted by Sam Ruby at

Why is this a good thing? Why should WHATWG follow this?

Posted by Shelley at

Btw, this ‘fix’ also allows inline XAML - perhaps the motivator? :)

Posted by Jeff Schiller at

Why is this a good thing? Why should WHATWG follow this?

Short version: I’d like to see the effective overlap between what can be done in HTML5 and XHTML5 increase.  It is the standards and technologies and capabilities that I am interested in, more so than the serialization format.

Longer version: I’d actually like to see this adopted, but outside of the TAG it has gotten zero traction.  Further background can be found here, which the WHATWG did act upon.

Posted by Sam Ruby at

Sounds vagely similar to XTF.

Posted by Minh Nguyen at

What I read in page 3 is even better: if you use a prefix for SVG and declare it on the “html” element, you can already display it in IE5 to IE7, provided that you use their “import” pseudo-processing-instruction.

In brief, I copy-pasted the code from page 3 and opened it in IE7, and it "just works"!

Far from perfect, but still good to know (I knew the trick with HTCs, didn’t know it with "object"s).

...now continuing my reading...

Posted by Thomas Broyer at

“It is the standards and technologies and capabilities that I am interested in, more so than the serialization format.”

I can agree on the overlap. But, to me, there’s more at stake here than a serialization format.

First of all, any of the browsers that unilaterally roll something out as de facto standard, tosses a few bits into the public domain (kind of), and then starts up another round of cross-browser implementation issues is not going to get an “attaboy” no matter how good the idea.

To be honest, I’m also not that fond of continuing the encouragement of bad markup. Sure XHTML doesn’t provide fun error messages. Is providing a version of HTML that says, “Go ahead, be sloppy” any better? To me this highlights a race to the bottom: crap happens, why fight it?

Second, MS will never support XHTML, or SVG, or MathML, or probably anything else that doesn’t suit it’s purpose. Some folks will say, why should it. I used to think because you’ll get slammed by developers and designers if you don’t — but that was before I met today’s developer and designer who doesn’t care as long as he or she gets the back button.

Already, I’m doing something that’s standard, serving up my pages as application/xhtml+xml. I have to change if I want my pages read by IE. I could be used to that if I weren’t so disgusted with the “standards” people who don’t see a problem with this.

Plus, excuse me for saying this, but you’re thinking like a hacker; that the gaps in what MS supports can be hacked around. I’m thinking like a person who knows I am going to continue for the next five years getting some IE dweeb telling me in my comments about how I’m breaking their browser, yet I’ve meticulously worked to ensure my pages are standard.

I don’t want to have to hack, or tell person after person “install this, install that” just because the stuff that should be implemented competes with Microsoft’s Silverlight crap.

Posted by Shelley at

Plus, excuse me for saying this, but you’re thinking like a hacker

That’s fair.

I don’t want to have to hack, or tell person after person “install this, install that”

I, too, don’t want to have to hack.

As I see it, the current status is: “your browser is broken, get a new one”.

With IE-8, the potential is there for “your browser is incomplete”.

Every time I install a new version of Ubuntu, I am reminded that Firefox, too, is incomplete — the first time I visit a site like YouTube.  Or Jacques' page.

Posted by Sam Ruby at

Sam
Maybe I forgot to trip some super-duper quirk mode but IE8 Beta 1 really screws up the layout of [link]

Posted by Dilip at

What’s on Jacques' page?

Posted by Robert Sayre at

What’s on Jacques' page?

I presume Sam means MathML, which requires separate fonts that are not bundled with Firefox (presumably for license reasons? I don’t really know).

Posted by Mark at

Mark gets it in one!  These fonts.

Now downloading IE-8 to see how standard compliant it really is.

Posted by Sam Ruby at

Your main blog page looks borked, but your individual post page is legible.

My fallback menu goes bye-bye in IE8’s standards mode (not sure why yet).  If I click the “Emulate IE7” button (naturally I have to “reboot” my browser for this), then the menu is back.

No link to submit bugs to Microsoft, that I can see...

Why am I wasting my time with this browser?

Posted by Jeff Schiller at

Also, apparently they don’t like the number 5: [link] in my ordered list of comments...

Posted by Jeff Schiller at

I presume Sam means MathML, which requires separate fonts that are not bundled with Firefox (presumably for license reasons? I don’t really know).

I don’t know either. Hopefully we can jump on the Web font bandwagon pretty quickly after Fx3. :)

Posted by Robert Sayre at

When I try the SVG example in IE8, I get an element with name SVG, with an attribute with name xmlns="http://www.w3.org/2000/svg"> <linearGradient (sic) and value http://www.w3.org/2000/svg"> <linearGradient, plus lots of similar attributes, and then LINEARGRADIENT and STOP and STOP and /LINEARGRADIENT elements. It looks like they’ve done something to HTML-namespaces in this build of IE8, since IE7 doesn’t do the crazy attribute thing, but it’s nothing like what it’s meant to be...

Posted by Philip Taylor at

Your main blog page looks borked, but your individual post page is legible.

My main page uses HTML5 elements and Sjoerd’s workaround.  My individual post pages use div.  Perhaps it is time to convert to all HTML5.  After all, it is not like the IE team is unaware of HTMLWG or anything...

No SVG by default.  HTTP headers via test.cgi:

HTTP_ACCEPTimage/gif, image/x-xbitmap, image/jpeg, image/pjpeg, application/msword, application/xaml+xml, application/vnd.ms-xpsdocument, application/x-ms-xbap, application/x-ms-application, application/x-shockwave-flash, application/x-silverlight, */*
HTTP_USER_AGENTMozilla/4.0 (compatible; MSIE 8.0; Windows NT 5.1; .NET CLR 1.1.4322; .NET CLR 2.0.50727; .NET CLR 3.0.04506.30; .NET CLR 3.0.04506.648; .NET CLR 3.5.21022)
HTTP_ACCEPT_LANGUAGEen-us
HTTP_ACCEPT_ENCODINGgzip, deflate

Posted by Sam Ruby at

The IE team posted on their blog about some kind of error reporting tool that can be used to complain about pages that don’t render correctly.  I submitted a report about Sam’s page for whatever good it might do.

Jeff:  The same post over at the IE blog mentions that you can submit bugs/feedback on their forums found here: [link]
Of course you probably need a Windows Live ID before you can post anything.

Posted by Dilip at

I presume Sam means MathML, which requires separate fonts that are not bundled with Firefox (presumably for license reasons? I don’t really know).

The beta version of the STIX fonts were released under this license. The final version (which, given the glacial pace at which they move, may be years away) seems likely to be released under a more liberal license.

Even with the STIX fonts, this page or this one will only render correctly in the latest Firefox nightlies (modulo this SVG bug, that I really ought to file a report on).

Posted by Jacques Distler at

Thanks Dilip - I have sent an email to Microsoft asking to be part of the Testing program...

Posted by Jeff Schiller at

Internet Explorer 8 lupaa paljon hyvää web-kehittäjille

Ilmeisesti ennen aikojaan on verkkoon vuotanut linkki sivustoon (via), joka esittelee IE 8:n uusia ominaisuuksia ja tarjoaa beta 1:n ladattavaksi. Lataus- ja lisätietolinkit eivät kirjoitushetkellä toimi (Päivitys: white paper -linkit alkoivat heti...

Excerpt from Aki Björklund at

Please also keep your hacks available/up-to-date on [link] for the people stuck with IE to find and use.

Posted by stelt at

By: sdodd

Unfortunately, neither IE6 nor IE7 support SVG. Actually, it appears at though SVG support in IE8 will be possible with binary behaviors . And generally speaking, if sites have demanded users install the Flash plugin in the past, why couldn’t they...

Excerpt from Comments on: 'I am not a Microsoft "fanboy"'... at

SVG and MathML Annexes to HTML5

As Anne previewed, HTML5 has recently added support for data attributes and MathML and SVG vocabularies. The former feels OPMLish to me.  I predict it won’t be long before we see escaped HTML inside data attributes “in order to... [more]

Trackback from Sam Ruby

at

Add your comment