Profile cover photo
Profile photo
Aaron Bradley
Knowledge graph strategist
Knowledge graph strategist


Post has attachment
With release of Google Podcasts, Google updates developer guidelines and expands tools for podcast publishers

On 19 June 2018 Google announced the release of the dedicated Google Podcasts app for Android ( They concurrently made a number of changes to their developer documentation for podcasts (call-out link) to reflect the availability of Google Podcasts, as well as providing support for direct link generation and podcast analytics.

Adding Google Podcasts to the mix

With the reveal of the dedicated Google Podcasts app for Android (, the documentation has been changed to reflect publishers' visibility in the app.

For example, the podcasts guide previously said "You can enable your podcast to appear in Google Search results along with individual episode descriptions...." Now the opening paragraph reads (emphasis added) "You can enable your podcast to appear in Google Search results and Google Podcasts, along with individual episode descriptions...."

Updating a podcast

A new section has been added on how to update a podcast, focusing chiefly how to employ Google Pubsubhubbub ( to ensure that a podcast update happens as soon as possible.

Removing a podcast

A new section has been added telling publishers how to remove a podcast or podcast episode.

Sharing direct links to Google Podcasts

A new section has been added which provides information on how to add "a button to your site with the Google Podcasts app logo that opens directly to your podcast in the [Google Podcasts Android] app."

To this end, a utility has been added to Google's Podcast Publisher Tools ( to generate a direct link to your podcast by entering a podcast feed URL. This page, first published in April 2018 (, was previously titled the "Podcast Search Preview". The preview functionality is retained, but the page now hosts both the search preview and the direct link generation functions.

Podcast analytics

A new section details how publishers can track what streaming or downloading of podcast episodes is originating from Google Podcasts.

Post has attachment
Google now using structured data to facilitate user-initiated play of media from search results or the Google Assistant

As per the call-out link, Google has added support for "Media Actions" which "enable users to initiate media content (e.g. songs, albums, movies) on content provider applications via Google Search and the Google Assistant."

Unsurprisingly the structured data Google relies upon to accomplish this are chiefly the hierarchy, as well as schemas pertinent to the implicated media types.

It also leverages as a "envelop" that contains all the data items from a publisher, provided in the form of a page-independent JSON-LD feed ( And there's a new validation tool for such data feeds (

Read more about it in the post, but there's lots about this newly-support content type that I haven't explored. For example the specs employ both the property and the JSON-LD @id key. Interestingly the latter does not need to be resolvable (does not need to be a working URL).

#schemaorg #jsonld #mediaactions #structureddata #google

Post has attachment
The Bing markup validator now supports JSON-LD

While support for JSON-LD is not yet reflected on Bing's structured data guidelines page (, their validator ( - sign-in required) does now support JSON-LD (as per the call-out image).

That's about all I can say confidently for now. I've not yet tested what happens if you provide Bing with invalid JSON-LD, and I don't know if this indicates that Bing is now able to generate rich snippets for any type based on JSON-LD (the last time I revisited the tracking table at I didn't see any changes for the URLs I tested).

One way or the other, certainly a good sign that Bing intends to make good on their promise to support JSON-LD.

Hat tip to Alexis Sanders, who pinged me about this on Twitter:

#jsonld #bing #schemaorg #structureddata

Post has attachment
In the community, "participation distribution is less unequal than

This an other observations on the "participation and interactions within the community behind" can be found in this great (just-published) paper from the European Semantic Web Conference (ESWC) 2018 (, by +Sami Kanza, +Alex Stolz, +Martin Hepp & +Elena Simperl.

What Does an Ontology Engineering Community Look Like? A Systematic Analysis of the Community

As per the abstract posted as the call-out image, the study also found that " seems to facilitate a form of collaboration that is friendly towards newcomers, whose concerns receive as much attention from the community as those of their longer-serving peers." I concur, and credit to the likes of +Dan Brickley, +Richard Wallis, +Martin Hepp and other principals and community leaders for creating such an environment.

#schemaorg #ontologies #engineering #github

Post has attachment
Google Rich Results Test now supports movies, TV shows and events

On 27 May 2018 Google added support for the types "TV and Movie" ( and "Event" (

I tested it out using the Google examples for Movie and Event, respectively, and the tool did indeed validate the code and provide a preview.

Movie - Code used
Movie - Rich Result Test preview

Event - Code used
Event - Rich Result Test preview

Live page with Event markup
Rich Result Test preview

Nothing much to note about the test or its results: works as it has for Job posting, Recipe and Course, just as one would expect.

In both cases only one result type was provided in the Search Results Preview.

Post has attachment
Changes made to Google's structured data guidelines for datasets, suggesting dataset rich results on are on the way

A number of changes were on 7 May 2018 to Google's feature guide for datasets (see call-out link). Here's the summary with my takeaways.

We're likely to see rich results for datasets soon

The prior version said that "Dataset markup is available for you to experiment with before it's released to general availability" and warned that, while you're able to use the Structured Data Testing Tool for validation, that you "won't, however, see your datasets appear in Search."

Today, however, the "paw" warning icon (class "dogfood:") has been replaced with a star icon (class "note"), and the note suggests that the roll-out of dataset rich results is imminent.

This feature is in pilot, and you may not see rich results for datasets yet. However, we recommend that you add dataset structured data to your site in preparation for new dataset features in Search results.

Another indication that this is moving out of (pre-)beta is that the "express interest" button has been removed.

The definition of "dataset" has broadened

Under the heading "What qualifies as a dataset?" - a list with items like "A table or a CSV file with some data" - there are two new items.

* An organized collection of tables
* Files relating to machine learning, such as trained parameters or neural network structure definitions

(That second addition is especially interesting, IMHO.)

DCAT is now supported

Under the new heading "Our approach to dataset discovery" the guidelines say that Google "can understand structured data in Web pages about datasets, using either Dataset markup, or equivalent structures represented in W3C's Data Catalog Vocabulary (DCAT) format". There follows an example in JSON-LD employing that remains unchanged, as well as new example in RDFa employing DCAT.

This section also newly states that "also exploring experimental support for structured data based on W3C CSVW, and expect to evolve and adapt our approach as best practices for dataset description emerge."

Cave ergo probat

That's Latin for "let the tester beware" - at least according to Google Translate. :)

A helpful new section "Known Errors and Warnings" lists "errors or warnings in Google's Structured Data Testing Tool and other validation systems" that you may encounter, telling you what you can safely ignore and providing further information on warnings that might not seem clear.

Structured data properties

I didn't see any changes to the properties listed for Dataset, DataCatalog or DataDownload. The only change here is organizational, where property specifications have been consolidated under the type to which each belong (previously they were organized thematically).

#stuctureddata #datasets #schemaorg #dcat

Post has attachment provides dataset for ClaimReview markup

In a blog post, +Dan Brickley welcomed "the new Data Commons initiative, which is devoted to sharing [structured data] datasets, beginning with a corpus of fact check data based on the ClaimReview markup as adopted by many fact checkers around the world."

This corpus, of select 8,902 ClaimReview instances from a small set of publishers (, Snopes, PolitiFact, The Washington Post, The Weekly Standard), is available as a download on the site (see below).

The data is provided as JSON-LD. I don't know whether or not this is simply pulled from the original source or whether any transformation has been applied (especially a syntactical transformation from microdata or RDFa to JSON-LD).

Call-out image is a Structured Data Testing Tool-triggered preview of one of the claims in the corpus, in this case from The Washington Post

Oh, and this dataset uses the brand-spanking new sdLicense property ("A license document that applies to this structured data, typically indicated by URL"):
Value for this in the code fueling the call-out image:

Post by +Dan Brickley on the blog and pages

Academia, Publishers and Tech Come Together to Open Up Fact Check Data

About FactChecks and Data

ClaimReview dataset download

Post has attachment
"You don’t need to use schema markup on your pages. You don’t even need to have a website to have a listing in Google Maps. But it’s probably smart to."

A great interview conducted by +Martha van Berkel with Community heavyweight +Bill Slawski, where Bill talks entities, Google patents, identifiers, structured data markup and everything in between. :)

Post has attachment
Google introduces preview tool for podcasts

Yesterday Google made a change to their feature guide page for podcasts ( Under the heading "Preview your podcast" the page says "Use the Podcast Preview Search Preview tool to see how your podcast may look in search results", and links to the preview page URL:

As per the call-out image, when a podcast (RSS) URL is preview-able the tool responds with a QR code (!), and instructs the user to scan "with your Android device to see the preview result." This QR code is generated whether or not you're using a desktop computer (middle left in the image) or an Android device (middle right).

The QR code itself is a deep link to the Google app (bottom left). Upon launching it generates a preview ID in the app:

I was unable to successfully generate a preview. When attempting to launch it from Google Assistant (bottom left) it's just treated like a search query string (result not pictured). When I launched it from a QR scan app (bottom middle) it crashes the Google App (bottom right).

Looking forward to seeing what this preview looks like. If anyone can get this to work please share your method and a screenshot! :)

#podcasts #rss #google #richresults

UPDATE - 27 April 2018 - Preview information and link to the preview page removed (see image in comments).

Post has attachment
Review rich results out of beta

Looks like publishers no longer need to register their interest to use critic reviews or review snippets ... so code away. :)
Wait while more posts are being loaded