Profile cover photo
Profile photo
Joe Pairman
Structured Information Architect
Structured Information Architect


Post has attachment
A post about using metadata and Linked Data with structured content; specifically DITA XML. It's based on the post I shared a few days ago, but I broke up the wall of text with some diagrams and graphics, and made the whole thing a bit more focused.
Add a comment...

Isn't Schema dot org more of a (light) ontology than a vocabulary?

(This might be more a question for the Schema dot org list, but I'm posting it here first because it might be a dumb question and members of this community are a very tolerant and friendly bunch 😉.)

Schema dot org is described as a vocabulary, rather than an ontology. Certainly, it doesn't provide a framework to provide rock-solid inferences in the way that a full-blown ontology would do. And yet, in other ways, it looks very much like a (light) ontology. Entities are subclasses of other entities. Each property may be permissible on one or more entities, but not others. For sure, there are a few lists of taxonomical values defined — things like . But not really that many.

I don't want to get hung up on the precise definition, but I wonder whether defining it and treating it as a light ontology (RDFS / RDFS+) and managing it with appropriate tools thereof might be useful, particularly as it continues to grow?

Hi all, can I check my understanding of the significance of SDTT "error" entries?

I believe that an error means
- if there's an associated feature , ( ) you won't get that.
- Google will still pick up on the rest of the data that is present apart from the error, and use that to inform normal (non-feature) search results

Does that sound right? Thanks for any tips.

(The reason I'm asking is that I'm evaluating a generator for a friend — one that seems to omit some Google-required properties and hence produces one or two errors in the SDDT. The reason I'm not yet recommending something more tried and tested is that the generator comes as part of a domain-specific site management bundle, and my friend's client really likes the rest of the offering. So I want to understand first how dim a view to take of these missing properties.)

Does Google care whether a JSON-LD script element is in the HTML head or body?

Just a quick one this time: does it make any difference whether an island of JSON-LD is in the HTML head element or the body? If it doesn't make any difference, I'd like to give my developer the option to put it in the body, which could well be easier in the CMS I'm working with.

Thanks in advance,

How can I raise a question/discussion point for the Schema․org Community Group? I want to understand the intention behind the separate types for TechArticle (which is described as being for "How-to (task) topics, step-by-step, procedural troubleshooting…") and HowTo itself.

(I can see that HowTo's structure implies procedures with steps whereas TechArticle doesn't have that. I assume that TechArticle is specifically for technical subjects whereas HowTo is broader. And I also see that TechArticle has dependencies and speakable properties, which would be very useful on a HowTo. I suppose that if one wanted steps and these properties one would create a multi type entity, but I'm still wondering why they're separate entities in the first place.)

Is there a more interactive way than the Google Form where I can raise this question to the community group? Is it alright to join the group directly and post away? Or is there a less formal way that interested outsiders such as myself should raise this kind of thing?

Structured Data Testing Tool: should I care that seemingly valid id "is not a known valid target type for the identifier property"?

I know that the SDTT isn't perfect and that various of its messages aren't worth worrying about. One that's just cropped up is that my ID values in JSON-LD (sourced from Schema App) are now reported as "not a known valid target type for the identifier property". See the recipe entity at:

This isn't the case for all my marked-up pages, however — the ID of the review entity here is fine:

Any thoughts on what might be different, and whether this is something to be concerned about?

What does Google do with duplicate entities on the same page?

I'm finally trying Schema App properly, using my blog as a vehicle. The blog's on the Squarespace platform, which is actually not bad but does put in its own meta tags. Even turning on developer mode and editing the templates doesn't help — if you omit the variable that brings the meta tags in, you lose huge swathes of functionality and might as well build the whole site from scratch.

Squarespace's meta tags on blog posts are particularly annoying. Posts are defined as articles, but the author, datePublished, headline, and publisher properties are all omitted. I can put all this in and more as JSON-LD via Schema App, but then Google sees two article entities (see ). How much of a problem is this? Is Google going to see the first entity and ignore the second, combine them, or something else?

Thanks for any tips.

Post has attachment
A nice marina in San Diego. Fuji X20, aperture priority, some mild tweaks in Apple Photos.
2 Photos - View album

Post has attachment
Structured Content Meets Taxonomy

The last couple of years, I've been working on some fascinating projects integrating standards-based taxonomies with structured content (primarily DITA XML). This free web session by +Andreas Blumauer and me discusses how to combine these two approaches. It's on June 7th. You'll learn about:
- Benefits and principles of granular structured content, and how to start preparing your own content for this new architecture.
- Best practices for linking structured content to standards-based taxonomies, and some pitfalls to avoid
- The underlying semantic architecture that you can work toward for a truly mature and scalable approach to linking content and data
- Key use cases that you can apply to your own organization

#taxonomy #structuredcontent #intelligentcontent #ux #standards #metadata #skos #ditaxml
Add a comment...

I'm informally researching for an upcoming event. Have most marketers (including content marketers) heard the term "structured content". And are they, on the whole, intrigued, indifferent, or horrified by the idea?

This could be any sense of "structured content" including markup, COPE-style approaches in a Web CMS, or hardcore XML stuff! Or just having heard the term and having no idea what it means.

#marketing #contentmarketing
Add a comment...
Wait while more posts are being loaded