Re: The Vocabulary, Schema.org governance, etc.

Hi elf,

> 1. Can search engines clearly list which vocabularies, other then
> schema.org, their support (or which subsets of them)? IMO collaboration
> with http://lov.okfn.org might come helpful.

I think this is something that should take place in the product forums of the respective search engine companies.

> 2. Do you have strategy on when to just support external vocabulary and
> when to integrate them into schema.org namespace? Example:
>   http://blog.schema.org/2012/11/good-relations-and-schemaorg.html

In the past, relevant external vocabularies have been integrated as a whole or in part into schema.org, see LRMI and GoodRelations.
As a rough guideline, I think a single contribution should not be more than 5-10 % of the current schema.org size - up to five new types is easy, upt to ten possible, 25 requires a lot of support from consumers, and bigger contributions have a very hard time.

In the future, we will need a fully functional extension mechanism so that bigger extension proposals can be properly hooked in without dominating the schema.org core namespace, IMO.

> 3. If some independent vocabulary gains broad adoption. How would
> process of adding support for it look? Especially if some Types overlap
> and properties collide. Examples:
>   * https://web-payments.org/specs/source/vocabs/commerce.html
>   *
> http://jasnell.github.io/w3c-socialwg-activitystreams/activitystreams2-vocabulary.html
> 

In the case of GoodRelations, we simply renamed existing elements so that they meet the naming convention of schema.org and don't collide with existing schema.org elements.

It is possible that a future version of GoodRelations will even deprecate the "old" local names in favor of the "new" local names in schema.org even in the original purl.org namespace, because it gets burdensome to deal with name differences. There will be owl:sameAs, owl:equivalentProperty, and owl:equivalentClass axioms available in the GoodRelations spec to implement backwards compatibility.

> I don't expect anyone having clear answers to any of those questions!
> 
At least I tried.

> On a process side I would suggest moving conversations fine tuning
> particular vocab terms to github issues, and just announce bigger
> changes here with invite to related issue/pull request.
> Besides keeping better track on them, this would lower volume of traffic
> on this list and allow more space for interesting conversations about
> environment of *WebSchemas*. Including schema.org, microformats,
> activitystrea.ms and many others listed in lov.okfn.org
> 
This may work for small bugfixes, but additions to the conceptual model (like to automotive extension) are hard to discuss solely on the basis of github pull requests, in my opinion. Also, we would exclude a lot of non-developers who will have important domain expertise.

Many domain experts will be only occasional contributors to schema.org but invaluable reviewers of extension proposals.

Martin

Received on Saturday, 20 September 2014 09:40:45 UTC