Frequently Asked Questions

What does this service do?

It checks your web page for RDFa and displays any data found there. It also compares your data against the published recommendations from major consumers/users of RDFa data, to see how well your data matches their requirements.

If your page doesn't raise any errors or warnings, it doesn't mean that your data is perfect — it just means that your errors may have been too ingenious for an automated tool to detect!!

Your service says my page is fine, but Google/Facebook/whatever can't see my data!

Some services may have accidental or intentional deviations from the RDFa standard in their parsers. This service does not attempt to replicate such bugs; nor can it, because I don't have insider information about how Google/Facebook/whatever parses RDFa.

This service simply parses RDFa according to the W3C recommendations and compares the end result to the schemas accepted by major consumers.

What's all this "widget://" stuff?

RDFa allows authors to include relative links in their HTML, like <a rel="next" href="page2">next</a>. If you use the file upload or code pasting facilities on this RDFa checker, then, for the purposes of resolving relative links, the service assumes that your document's base URL is something like widget://blahblahblah.rdfa/self in the absence of evidence to the contrary.

If you're checking a web page by typing in its address or clicking on a link, then you should never see this stuff.

My page used to have no errors, but now it has lots!

That's because we're adding extra checks all the time!

How can I stop you checking my page?

This service respects the Robots Exclusion Standard (robots.txt), with a User-Agent name of "check.rdfa".

Alternatively, you could filter HTTP requests containing an HTTP User-Agent request header beginning with "check.rdfa".

Can't you ignore my robots.txt file?

No, but you can add this to it:

    User-agent: check.rdfa
    Disallow:

(i.e. disallow nothing - allow everything)

Why doesn't the checker output valid XHTML?

This service outputs valid XHTML+RDFa 1.1 and HTML5+RDFa. Most pages content-negotiate between the two. HTML5+RDFa is currently only a draft specification and is not yet supported by the W3C validator, which is why you may see some validation warnings if you try to validate check.rdfa.info pages.

Is there an API?

The checker can output data in JSON, YAML or eval-uable Perl. You can either add a parameter to the request along the lines of format=json, format=yaml or format=pl or include application/json, text/x-yaml or text/x-perl in the request's HTTP Accept header.

There's also support for PHP's serialize/unserialize format, but this is broken due to a known bug in a third-party library.

For completeness, format=html and format=xhtml work; as do the media types text/html and application/xhtml+xml.

<insert some sort of question about HTTP Referer>

Yes, http://check.rdfa.info/check/referer works how you want it to.

What software powers this service?

Perl. Here's a page listing the key Perl modules used along with which version is currently installled on the server.

When will you support Microdata/Microformats?

Never.

The aim of this tool is to check RDFa, and check it well. Adding support for other formats would distract from that aim.

The Perl modules underlying this site are available on CPAN. It would not be rocket science to pull out the RDFa parser and plug in a different parser instead. (And indeed, I have written suitable Microformat and Microdata parsers which are available on CPAN too.)

Where can report bugs?

Preferably via the Google Code project, though if you'd rather not make the details of your pages public, you can contact me.