htmlschema.orgsemantic-markupmicrodata

The correct approach to markup "keywords" of a blog post by using HTML5 and Microdata?


The keywords property from Schema.org, as it being in plural form suggests, seems to imply its value should be a line of comma-separated text. Like the following example:

<p itemprop="keywords">lorem, ipsum, dolor, sit, amet</p>

That reminds me of the <meta name="keywords" content="lorem, ipsum, dolor, sit, amet"> tag we used to put within <head></head> for SEO reason.

However, from semantics' perspective, I think the above example is incorrect because keywords should be a list of words. Thus it should be markuped by using the <ul> element.

Is there a common consensus on how the keywords of a blog post should be markuped by using HTML5 and Microdata?


Solution

  • Update: The definition of Schema.org’s keywords property changed. Now it makes clear that it expects multiple tags, and that they should typically be comma-separated.

    tl;dr: If you want to use a ul element for your tags, using Schema.org’s keywords property could look like this in Microdata:

    <article itemscope itemtype="http://schema.org/BlogPosting">
    
      <footer>
        <ul itemprop="keywords">
          <li><a href="/tags/foo" rel="tag">foo</a>,</li>
          <li><a href="/tags/bar" rel="tag">bar</a></li>
        </ul>
      </footer>
    
    </article>
    

    If using it like that, you have to make sure that the ul contains no other non-tag text.


    HTML

    The meta element with the name value keywords offers a way to specify keywords that apply to the whole page. These keywords consist of text only, so you can’t use URIs.

    So for the typical blog tags, you’d go the following way (possibly in addition to meta-keywords).

    Link type

    HTML5 defines the link type tag:

    The tag keyword indicates that the tag that the referenced document represents applies to the current document.

    Note that this link type can only be used on single post pages as the tag always applies to the whole page.

    (Attention: tag is a Microformat, too, but it has a different definition: You may only use specially crafted URLs for the Microformat tag.)

    If your tags are more like categories (controlled vocabulary instead of free tagging), you could use the link type category (possibly in combination with tag).

    Markup for the tags

    You might use a ul or a dl:

    <ul>
      <li><a href="/tags/foo" rel="tag">foo</a></li>
      <li><a href="/tags/bar" rel="tag">bar</a></li>
    </ul>
    
    <dl>
      <dt>Tags</dt>
      <dd><a href="/tags/foo" rel="tag">foo</a></dd>
      <dd><a href="/tags/bar" rel="tag">bar</a></dd>
    </dl>
    

    I’d go with the dl when you also have other metadata to list, e.g. the author, publication date, etc.

    A div with comma-separated a elements would work also, of course:

    <div>Tags: <a href="/tags/foo" rel="tag">foo</a>, <a href="/tags/bar" rel="tag">bar</a></div>
    

    Container for the list

    The list should be part of a footer element (inside the article if you use one for your blog post):

    A footer typically contains information about its section such as who wrote it, links to related documents, copyright data, and the like.

    Vocabularies

    Schema.org

    CreativeWork defines the property keywords (BlogPosting inherits it):

    Keywords or tags used to describe this content. Multiple entries in a keywords list are typically delimited by commas.

    It expects plain text, so there’s no way to provide the tag URLs in Schema.org.

    As keywords expects a comma-separated list of tags, dl can’t be used (unless the dl contains nothing else than tags).

    With an ul, it could look like this in Microdata:

    <ul itemprop="keywords">
      <li><a href="/tags/foo" rel="tag">foo</a>,</li>
      <li><a href="/tags/bar" rel="tag">bar</a></li>
    </ul>
    

    If using a div, you’d need to add another div/child so that the label "Tags:" gets not interpreted as tag itself:

    <div>Tags: <span itemprop="keywords"><a href="/tags/foo" rel="tag">foo</a>, <a href="/tags/bar" rel="tag">bar</a></span></div>
    

    Common Tag

    Common Tag is a (RDF) vocabulary for tagging. In contrast to Schema.org’s keywords property it uses URIs for tags, not text only.

    Example from their Quick Start Guide in RDFa:

    <div xmlns:ctag="http://commontag.org/ns#" rel="ctag:tagged">
       NASA's <a typeof="ctag:Tag" 
                 rel="ctag:means"
                 href="http://rdf.freebase.com/ns/en.phoenix_mars_mission"
                 property="ctag:label">Phoenix Mars Lander</a> has deployed its robotic arm. 
    </div>
    

    Explanation: The content is tagged (ctag:tagged) with a tag (ctag:Tag). This tag is defined (ctag:means) by the URI "http://rdf.freebase.com/ns/en.phoenix_mars_mission" and it’s called (ctag:label) "Phoenix Mars Lander"

    Instead of ctag:Tag you could use the class ctag:AuthorTag (which means that it’s tagged by the content author).

    MOAT

    There is also the MOAT vocabulary, which stands for "Meaning Of A Tag". Unfortunately, their website seems to be gone (?).

    An ontology that let users define relationships between Tag objects and URIs of Semantic Web resources