Re: Historical events

Hi Roger !

Good points.  However...

It is all up to the consuming applications to extract value from any
structured data you provide.  Google, Bing, MyApplications, etc.

99% of the applications will understand what you mean when you say

    Event : Bronze Age

If a machine or application has difficulty understanding the Semantic
meaning of that statement ?  Then you can supply just a small handful of
additional structure to tell that machine or application what you really
mean.

    Event : Bronze Age
    sameAs : https://www.wikidata.org/wiki/Q11761

Your looking at the Event class description with the context of "...such
as..."  DON'T, we provide merely examples.  It is best to concern yourself
with only the leading description...
"an event happening at a certain time and location"

And where we already have an understanding that Location is ... OPTIONAL :)

In fact, in all things Schema.org and Semantic Web, the only must-have
property rule is that of "name".  Everything else is bonus structured data.
"Some structure is better than no structure at all"

Its very likely in the next release that we update the description to
simply be ... (which in my opinion we should have done long ago)

"an event happening at a certain time, this could include a location when
necessary. Examples, concert, lecture, festival, Bronze Age"

-Thad

On Wed, May 30, 2018 at 12:25 PM Roger Rohrbach <roger@ecstatic.com> wrote:

> Thad,
>
>  The Event class’ description is “an event happening at a certain time and
> location, such as a concert, lecture or festival.”  Its
> properties—composer, performer, organizer, audience, doorTime,
> workPerformed, review etc.—clearly adhere to this narrow definition.
> You’re suggesting that we shoehorn World War II battles and the Bronze Age
> into this schema, and quell any uneasiness thus induced by referencing an
> external vocabulary.  I will admit that this is possible; to my mind, it is
> undesirable.  Were the Event type truly a “general type” capable of serving
> as the superclass for these two different semantic elements, I’d feel
> differently.
>
> To use an analogy, it’s as if  schema.org provided Funeral, but not
> Ceremony, and you told me “just use Funeral to represent Wedding, and
> reference  https://www.wikidata.org/wiki/Q49836 as the additionalType.”
>
> I accept the premise that this is not a universal ontology.  But what
> classes there are, ought to retain their semantics.  The Bronze Age is not
> a-kind-of concert, lecture or festival.  I can’t see how your approach
> would result in truly machine-readable content.
>
> respectfully,
>
> Roger
>
>
> On May 29, 2018, at 2:13 PM, Thad Guidry <thadguidry@gmail.com> wrote:
>
> Allison and Roger... see examples below.  Yes, you can still use Event.
> When you don't have a startDate or endDate, then just leave them excluded.
> The use of additionalType and referencing other ontologies or Wikidata is
> quite useful and a generally accepted best practice when you need to easily
> subtype things that Schema.org has only general types available.
>
> {
>   "@context": "http://schema.org",
>   "@type": "Event",
>   "sameAs": "https://www.wikidata.org/wiki/Q327052",
>   "additionalType": "https://www.wikidata.org/wiki/Q15815670",
>   "name": "Battle of Aachen",
>   "startDate": "2 October 1944",
>   "description": "major conflict during World War II",
>   "endDate": "21 October 1944"
> }
>
> {
>   "@context": "http://schema.org",
>   "@type": "Event",
>   "sameAs": "https://www.wikidata.org/wiki/Q11761",
>   "additionalType": "https://www.wikidata.org/wiki/Q15401699",
>   "name": "Bronze Age",
>   "description": "prehistoric period",
>   "about": "Historical Event"
> }
>
> Any other questions ?
> -Thad
>
>
>