Can I add Microdata from HTML5 to a XHTML Strict site and still be compliant?

StackOverflow https://stackoverflow.com/questions/3659778

Question

I've got a site coded in XHTML 1.0 Strict. I want to use the new Microdata to add breadcrumbs to my site (so Google will understand them).

My old non-microdata marked-up breadcrumbs look like this:

<ul>
  <li><a href="...">Level 1</a></li>
  <li><a href="...">Level 2</a></li>
  <li><a href="...">Level 3</a></li>
</ul>

According to Google, to markup breadcrumbs using Microdata, you extend the above code like this:

<ul>
  <li itemscope itemtype="http://data-vocabulary.org/Breadcrumb">
    <a href="..." itemprop="url">
      <span itemprop="title">Level 1</span>
    </a>
  </li>
  ...
</ul>

But this is not valid XHTML 1.0 Strict.

What should I do?
Should I ignore the validation conflicts?
Should I write itemscope="itemscope" instead of just itemscope (this would be valid XML, but still not valid XHTML)?
Should I change the Doctype to be HTML5 instead of XHTML 1.0 Strict?

I want this to work all the way back to IE6!

Please advice :)

Was it helpful?

Solution

Yes, if you wanted to use itemscope in XHTML, you would need to write itemscope="itemscope" and use XHTML5 (same DOCTYPE as HTML5, but XML syntax).

itemscope is not included in W3 HTML5, but present in WHATWG's version, so validation may continue to be a difficulty. There seems to be quite some political argument on this issue, which I haven't been following as it looks fairly tedious.

For the moment, if you want to use breadcrumb annotations in a finalised, validatable document format, you could use RDFa instead: the alternative (but older) proposal, which the argument is all about, and use the existing doctype:

<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML+RDFa 1.0//EN" "http://www.w3.org/MarkUp/DTD/xhtml-rdfa-1.dtd">

OTHER TIPS

Since the major search engines have decided on schema.org last June (2011) as the way to do rich snippets, this question has become much more important seeing XHTML5 does not yet have a working DTD (BTW, http://www.html5dtd.org/ is working on a XHTML5 DTD and may well be ready when you read this, if so disregard what I'm about to say). And what I am about to say summarises a page I placed at http://www.nedprod.com/programs/portable/XHTMLwithHTML5microdata/ a few weeks ago, and there has more detail including a rich snippets demo if you want it.

I had need of extending XHTML 1.x Strict with schema.org/HTML5 microdata and getting it all to validate properly for updating nedprod, and Microsoft Expression Web has the occasional tendency to eat bits of HTML it edits, so validation is handy for catching when it borks. Hence I have created these DTDs which extend the standard XHTML 1.0 ones:

To use, take a copy of your desired DTD (don't use the original from nedprod, I can't afford the bandwidth) and use as follows:

<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict with HTML5 microdata//EN" "xhtml1-strict-with-html5-microdata.dtd">

or ...

<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional with HTML5 microdata//EN" "xhtml1-transitional-with-html5-microdata.dtd">

... or more likely, override the DTD used for validation by your particular XML validating setup.

BTW, here's something interesting, and I only include this as it's useful to know when answering the question. I honest to God thought that using the above doctypes would invoke quirks mode when rendering. Turns out, much to my great surprise, that IE8, Chrome 14, Firefox 5 and Opera 11.50 all render such a doctype in Standards mode. Who would have thought! So you could, if you wanted to, upload your XHTML pages onto the public internet with the custom doctype and the newer browsers at least would do the right thing.

Hope this helps someone,
Niall

Valid HTML 5 example

<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>Example page</title>
</head>
<body>
    <div itemscope="" itemtype="http://schema.org/MediaObject">
        <div itemprop="video" itemscope="" itemtype="http://schema.org/VideoObject">
           <meta itemprop="name" content="Breast Augmentation Video Diary">
           <meta itemprop="duration" content="PT12M54S">
           <meta itemprop="thumbnailUrl" content="http://www.plastic-surgery-estonia.com/new-assets/images/thumbnails/breast-augmentation.jpg">
           <meta itemprop="contentURL" content="http://www.youtube.com/watch?v=BwPN6eCpxTk">
           <meta itemprop="embedURL" content="http://www.youtube.com/watch?feature=player_embedded&amp;v=BwPN6eCpxTk">
           <meta itemprop="uploadDate" content="2010-11-09">
           <iframe src="http://www.youtube.com/embed/BwPN6eCpxTk?rel=0&amp;autohide=1&amp;modestbranding=1&amp;showinfo=0"></iframe>
           <span itemprop="description">Video Diary</span>
        </div>
   </div>
</body>
</html>

Found these links helpful:
- http://support.google.com/webmasters/bin/answer.py?hl=en&answer=2413309&topic=1088474&ctx=topic
- http://www.reelseo.com/embedded-youtube-indexed-google/

Use the application/ld+json MIME type and a microdata generator to transform the markup into data:

<!DOCTYPE html>
    <html xmlns="http://www.w3.org/1999/xhtml">
    <head>
    <title>microdata.xhtml</title>
    <meta http-equiv="Content-Type" content="application/xhtml+xml; charset=utf-8"/>
    </head>
    <body>
      <div>
      <script type="application/ld+json">
      {"items": [{
        "type": ["https://schema.org/breadcrumb"],
        "properties":{ 
        "url": ["..."],
        "title": ["Level 1"]
        }
       }]
      }
      </script>
      </div>
    </body>
    </html>

or a data:uri in an object tag:

    <!DOCTYPE html>
    <html xmlns="http://www.w3.org/1999/xhtml">
    <head>
    <title>microdata.xhtml</title>
    <meta http-equiv="Content-Type" content="application/xhtml+xml; charset=utf-8"/>
    </head>
    <body>
      <div>
        <object data="data:text/html;charset=utf-8;base64,PHVsPiA8bGkgaXRlbXNjb3BlIGl0ZW10eXBlPSJodHRwOi8vZGF0YS12b2NhYnVsYXJ5Lm9yZy9CcmVhZGNydW1iIj4gICAgICAgICA8YSBocmVmPSIuLi4iIGl0ZW1wcm9wPSJ1cmwiPiAgICAgICAgICAgPHNwYW4gaXRlbXByb3A9InRpdGxlIj5MZXZlbCAxPC9zcGFuPiAgICAgICAgIDwvYT4gICAgICAgICA8L2xpPiAgICAgICA8L3VsPg==">

          <?microdata
          <ul>
            <li itemscope itemtype="http://data-vocabulary.org/Breadcrumb">
             <a href="..." itemprop="url">
               <span itemprop="title">Level 1</span>
             </a>
            </li>
          </ul>
          ?>

        </object>
      
        <!--[if lt IE 8]>
        <object data="mhtml://#foo">

          <?microdata
          <ul>
            <li itemscope itemtype="http://data-vocabulary.org/Breadcrumb">
             <a href="..." itemprop="url">
               <span itemprop="title">Level 1</span>
             </a>
            </li>
          </ul>
          ?>
          <div id="foo">
              PHVsPiA8bGkgaXRlbXNjb3BlIGl0ZW10eXBlPSJodHRwOi8vZGF0YS12b2NhYnVsYXJ5Lm9yZy9CcmVhZGNydW1iIj4gICAgICAgICA8YSBocmVmPSIuLi4iIGl0ZW1wcm9wPSJ1cmwiPiAgICAgICAgICAgPHNwYW4gaXRlbXByb3A9InRpdGxlIj5MZXZlbCAxPC9zcGFuPiAgICAgICAgIDwvYT4gICAgICAgICA8L2xpPiAgICAgICA8L3VsPg==
          </div>
       </object>
       <![endif]-->
      </div>
    </body>
    </html>

I want this to work all the way back to IE6!

Use the application/xhtml+xml XSLT shim to support IE6 and extend it to get a copy of the markup.

References

try and validate some of google's pages...they don't validate. validation is a tool, an awesome one, but nothing more, although i do applaud your determination. if you're that worried about validation i would switch to HTML5 .

Licensed under: CC-BY-SA with attribution
Not affiliated with StackOverflow
scroll top