I maintain a WordPress plugin that allows authors to makes widgets representing other pages on their sites with a linked title, thumbnail image, and excerpt of the text. I've used hAtom to mark up the entry, title, bookmark, and summary but feel that displaying the author and updated date are nonessential, even undesireable because the pages are static, authorless content.

Here's the reduced test-case HTML:

<article class="hentry">
    <div class="entry-title">
        <a href="http://example.org" rel="bookmark">Who we are...</a>
        <img src="http://example.org/img.jpg" />
    <div class="entry-summary">
        <p>Summary of the Page</p>

With the Rich Snippet Testing Tool or webmaster tools, it extracts all the expected data but throws errors:

Error: Missing required field "updated".

Error: Missing required hCard "author".

I know that both updated and author are required according to the hAtom spec.

I'm fairly confident that this doesn't hurt the site's SEO or anything else, I'm trying to understand the various trade-offs here. I could:

  1. Remove the hAtom markup completely (possibly replacing it with schema.org markup?)
  2. Add visually hidden author and date info.
  3. Leave broken with the errors.

If an hAtom entry is truly invalid without updated and author, what are the concrete downsides and are there any ways to avoid them? If there are none, is there any way to suppress the errors? As I distribute this markup in a plugin, I'd like to avoid the errors if at all possible.

1 Answer 1


Invalid rich data markup

Having invalid failing rich data markup such as Schema or hAtom isn't going to have a massive impact on your rankings, in fact there's little to none evidence supporting that having good markup actually increases your rankings directly, however it is agreed that having rich data can improve the click rate because of additionally information being provided, especially with review stars for example.

Satisfying hAtom

If you want to validate your current markup then as you know you will need to use author and updated, if you do not want these elements visible then as long as these elements are triggable i.e viewable by both your visitors and search engines then you have nothing to worry about. A typical example can be found below:


<article class="hentry">
    <header class="entry-title">
        <a href="http://example.org" rel="bookmark">Article Title</a>
    <div class="entry-summary">
        <p>First Paragraph ideal for a introduction of the article.</p>
    <div class="entry-content">
        <p>The rest of the content.</p>
        <aside><img src="http://example.org/img.jpg" /></aside>
        <p>Published by <span class="author vcard"><span class="fn">John</span></span> on <time class="published" datetime="2013-06-13 12:00:00">13<sup>th</sup> June 2013</time> Last Update:<time class="updated" datetime="2014-03-14 12:00:00">Match 2014</time></p>


Simply using a hover over on the title of your article can easily reveal additional information, ideally however you should give some clue that this information can be viewed for user experience through not required, something like:

article footer{display:none;}
article header a:hover footer{display:block;}

Using Schema

You are correct that with Schema you will not need to include published time, date and updated nor author.

Simply using something like:

<article itemscope itemtype="http://schema.org/Article">
    <a href="http://example.org" itemprop="url">
        <span itemprop="name">Article Title</span>
    <div class="content-container">
        <p>Content to go here...</p>
        <aside><img src="janedoe.jpg" itemprop="image" /></aside>
  • Accepting mostly for this part "Having invalid failing rich data markup such as Schema or hAtom isn't going to have a massive impact on your rankings..." The display:none stuff would be inaccessible and might actually cause some SEO issues (purely conjecture) now that Google is starting to process JS & CSS. I'll probably switch to schema.
    – mrwweb
    Commented Nov 4, 2014 at 18:50
  • Display:none is absolutely fine as long as there is either a JS or CSS solution... See part CSS how easily this can be done, its very common for a lot of web elements to use hover overs (like movie covers with descriptions on hover, no reason why you should worry about using it for author information. Commented Nov 4, 2014 at 18:52
  • Screen readers ignore display: none so that's not an accessible solution for hiding content.
    – mrwweb
    Commented Nov 4, 2014 at 19:08
  • Well yes because screen readers are designed for impaired users and rightly so a element that normally has a hover over should be displayed in plain sight. Commented Nov 4, 2014 at 21:42

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.