Re: Historical events

An argument against a separate type for "historical events" is that as time goes by, you would have to update the type membership of an entity, though it's identity will not have changed. Image the ESWC 2018 conference: It is now a future event, starting on Sunday, but in a week from now, it will be a "historical" event. Still, most of the characteristics of the event, like startdate, enddate, location, sub-events, papers presented, etc. will not change.

One could consider being historical or ongoing/future as *roles* that an entity of type event may carry, same as a human being will be a child vs. a grown-up depending of the temporal context without changing its membership in the rigid type "human being".

Practically: Indicating being historical or non-historical/ongoing/future for an event can be covered very well using xsd:datetime literals. My ZX81 thirty-six years ago could have computed whether the event was historical or not from such data.

Best
Martin

-----------------------------------
martin hepp  http://www.heppnetz.de
mhepp@computer.org          @mfhepp




> On 01 Jun 2018, at 01:50, shawn fielding <shawnmfielding@gmail.com> wrote:
> 
> I am brand spanking new to this listserv, and pretty new to the concept of schema.org altogether, so I might be breaking protocol.  Please let me know if I am.  Don't want to do that.
> 
> But the idea of historical event as a subcategory of event seems to make a lot of sense to me in what I do.  I am not looking at general events, though being able to say "World War II" would be handy.  I am looking at things like when a ship was launched or when it sank as historical events.  (Would you be able to nest events?  Like event WWII -> event "sinking"?)  Right now event does not seem to fit this sort of thing.  I actually and trying to put together and propose a small, er, taxonomy(?) for vessels under vehicle. Seems to fit, almost.  One of the ideas I was considering was "date sunk."  But maybe that defeats the purpose of event itself.   Maybe it doesn't if I can't relate the ship to its historical event in the current structure.  I'm not sure how that works.  There are obviously people much more in the know here, so I am just throwing out these comments and questions to get greater clarity of how we can use schema.org to represent what we do effectively.
> 
> Shawn Fielding, novice 
> 
> 
> 
> On Thu, May 31, 2018 at 6:01 PM, Vicki Tardif <vtardif@google.com> wrote:
> One thing to consider is that if only the properties of schema.org/Thing are provided, the more important it is to have an accurate type. If I have
> 
> name: Ramones
> image: http://www.ramones.com/image.jpg
> url: http://www.ramones.com
> 
> I do not not know if this is the band, the self-titled album, or a book about the band.. I need the type to disambiguate.
> 
> And yes, one may use sameAs, but not everything is notable enough to be in Wikidata/Wikipedia.
> 
> - Vicki
> 
> 
> On Thu, May 31, 2018 at 5:33 PM Thad Guidry <thadguidry@gmail.com> wrote:
> Peter,
> 
> You used a string, not an identifier.  I am saying, when practical, use an identifier.
> 
> I will continue to use Event myself for my applications, it fits my needs quite nicely.
> 
> I get the feeling that the community tries to push back against someone who wants to roll their own way.  I was simply expressing how I roll my own way and that it works for me.
> 
> I am not asking for anyone to agree with me.  I don't care about agreement in this case because I am already deriving value from this usage for my own applications.
> 
> But I do understand that others on this list have very different needs than me, especially around SEO, and not interoperability concerns..  That's fine.  We all have different uses cases.  Which is why I was asking about Roger's use case.  Being able to categorize things into structure is great !  But how it gets used is what I am always most interested in.
> 
> -Thad
> 
>

Received on Friday, 1 June 2018 12:47:45 UTC